Ethereum’s All Core Developer calls can be difficult to follow, so this “Checkpoint” series aims for periodic high-level updates, based on what’s happening in core development. See previous update here.
tl;dr
Glamsterdam’s progress is slow but steady: implementing developer-builder separation (ePBS) is proving trickier than expected and non-flagship features such as gas repricing have their own complexities to resolve.
The next upgrade, Hegotá, has now chosen its major feature, FOCIL, with a commitment to work on account abstraction as part of the minor feature set. Non-flagship EIP proposals can be submitted starting April 9.
Glamsterdam
Implementation of Ethereum’s next upgrade, Glamsterdam, is well underway. The full list of planned and planned features and their summaries are available on Forkcast. Planned Features (“SFI”) will almost certainly be in the upgrade when it goes live. Planned Features (“CFIs”) are very likely to be included, but their inclusion will depend on their complexity to implement as well as the progress and compatibility of planned features being implemented before them. The definitions of these steps are subject to refinement to better describe the process!
Glamsterdam’s specific feature set includes several gas repricings that are mostly prioritized as a package. Ecosystem developers have also constructively expressed the importance of EIP-7954, an increase in the maximum contract size, so it will likely be a priority.
One of the main sticking points in Glamsterdam’s progress is ePBS, which splits block production into two parties acting in sequence under consensus instead of how it currently happens outside the protocol, so the protocol must now handle disagreements or failures between them. Each part of the stack must reason about “partial blocks” and bipartisan coordination, a change that affects virtually everything.
On the execution layer side, Block Level Access Lists (BALs) represent a fundamental overhaul of how gas and state access works, and BAL development networks are making predictable progress through these potentially difficult problems.
The developers are targeting Glamsterdam’s first widespread devnet next week, if the current ePBS devnet can be stabilized. Next, they will go through several such devnets that include more and more non-flagship features. Once they have a stable devnet with all the features that will be included, they will remove the client builds, do the final security reviews, move to the testnets and, once these are confirmed stable, finally announce the mainnet fork date.
Hegotá
Hegotá is the upgrade after Glamsterdam – the selection process for its main feature is complete, with FOCIL (EIP-7805) selected as the consensus layer headliner. The execution layer had a heated debate over an account abstraction proposal, but the lack of consensus on specific implementation choices led to the proposal (EIP-8141: framework transactions) being moved to Considered for Inclusion (CFI) status as a lead, a reserved commitment to work on an account abstraction proposal that garners broader support among client developers, with community participation.
There is growing interest in prioritizing quantum resistance, but no standalone proposals have so far been introduced – only abstraction proposals that consider quantum resistance as part of a larger whole.
Starting April 9, anyone can propose a non-flagship feature for Hegotá by placing the EIP in the Proposed for Inclusion section of the fork’s EIP meta. The deadline for doing so will be announced at least two weeks in advance. You can always find the most up-to-date information about the Hegotá process and schedule on the Hegotá forkcast page.
The timing of this upgrade will heavily depend on the progress Glamsterdam makes over the coming months.
Gas limit
Gas limit increases are continually tested on devnets, with the goal of enabling higher limits in Glamsterdam and beyond. The current target for the gas reference limit is 60 M, but testing is underway at much higher limits to understand the implications and optimizations needed to achieve these higher limits. Much of the price review work currently being implemented in Glamsterdam will achieve higher, safer limits.
PEI Champion Handbook
New resources have been published on the Protocol Support Team website to help authors and advocates defend feature proposals and solicit feedback from stakeholders. If you want to integrate your feature into an Ethereum upgrade, visit the EIP Champions Handbook to navigate the process.
Fusaka’s speed in following Pectra sparked enthusiasm for faster Ethereum forks, but Glamsterdam proved trickier and slower than expected. The feature changes are far from trivial, although client developers don’t seem particularly daunted by the challenge. Glamsterdam in Q2 seems unlikely to me, but the sole headliner chosen for Hegotá means that, depending on how the non-headlining process goes, he could follow Glamsterdam in a shorter time frame than Glamsterdam follows Fusaka. It is always possible that client developers will converge on an account abstraction proposal and promote it to the forefront in recognition of its importance to the community.
While not an “official Ethereum roadmap,” the roadmap released this year has reignited enthusiasm for having a guideline that can help inform feature selection during the upgrade process and so far appears to be a beneficial tool to help the community agree on priorities.
Relevant ACD calls:
(January 21 – April 9)
ACDT: 76, 75, 74, 73, 72, 71, 70, 69, 68, 67
ACDC: 176, 175, 174, 173
ACDE: 234, 233, 232, 231, 230, 229
